Personalized Recommendation System for Advanced Learning Management Systems

The information on the web is ever increasing and it is becoming difficult for students to find appropriate information or relevant learning material to satisfy their needs. Machine Learning (ML) and Data Mining (DM) have emerged in a variety of application areas including in Learning Management Systems (LMS). In the learning field, the main focus is on the learning style and learning behavior of the learners. Identifying learning style and learning behavior helps in the development of learning management systems. Effective Personalized Learning Recommendation Systems will not only reduce this burden of information overload by recommending the relevant learning material of their interest to the students, but also provide them with the "right" information at the "right" time and in the "right" way. Educational Data Mining is an emerging interdisciplinary research area of DM that deals with the development of methods to explore data originating in an educational context. In this paper, we present a novel technique for finding the relevant references, i.e., Most Recently Referred (MRR) and All Time Referred (ATR) titles by students in LMS. The MRR references are obtained using a personalized dynamic sliding window, which is able to adapt its size according to the ratio of references/titles mentioned by students' in the previous semester. The ATR references are obtained by selecting references that represent the interest of a larger number of students in a particular reference over the year(s). This novel approach has helped in incrementally updating the association rules mined from the log files of an LMS database. The experiments and the evaluation of the proposed methods show that the MRR and ATR referred titles are in sync in numbers and, hence, we can explicitly recommend the learning material references by using either of the proposed techniques.